The Equilibrium and Kinetic Properties of the Aqueous Hydroxide

At 298°K klz = 0.55 =k 0.01 M-l sec-l. For the equilibrium kza. (NC)6Fen'0~4-. $ HzO. (NC)sFeH2O3- + NO$-. (2) k82. AH' = $12.3 =k 0.2 kcal/mole, ASo...
1 downloads 0 Views 455KB Size
PROPERTIES OF THE AQUEOUS HYDROXIDE-NITROPRUSSIDE SYSTEM573

Vol. 5 , No. 4, April 1966

found not to fit the same linear AGO-ASQ relation found for other carboxylic acids.% The two points for [1,12-Bl~Hl~(COOH)~]2on a plot of AGO vs. A S o fall on a parallel line with a different intercept. It would now be of interest to study other boranocarboxylic acids to learn if this intercept is characteristic of the AH" values for ionization of carboxylic acids with the carboxyl group attached to a boron atom.

ported. I n fact, no values calculated in the above fashion have been reported, probably because of the difficulty of determining r for a nonrigid structure. It is interesting that the ( b In D / b T ) value obtained, -2.5 x 10-8 deg-I, is of the same order of magnitude deg-l. as that reported" for bulk water, -4.7 X The thermodynamic values determined in this study for [1,12-BlzHlo(COOH)~]2-proton ionization were

CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY, UNIVERSITY OF CALIFORNIA, DAVIS,CALIFORNIA

The Equilibrium and Kinetic Properties of the Aqueous Hydroxide-Nitroprusside System BY JAMES H. SWINEHART AND PETER A. ROCK Received October 20, 1965 A complete thermodynamic and kinetic investigation of the aqueous hydroxide-pentacyanonitrosylferrate(11) (nitroprusside), (NC)5FeNOZ-,system has been carried out. For the equilibrium

+ 20H-

+ HzO

(NC)6FeNOZ-

= (NC)&FeN0z4-

(1)

A H o = -16.2 =k 0.9 kcal/mole and AS" = -26.1 i: 3.0 gibbs/mole. The equilibrium constant, K ~ za ,t 298°K is (1.5 =k 0.3) X 106. The mechanism postulated for the nitroprusside-hydroxide reaction is kii

+ OH- -+(NC)6FeNOzH3( NC)&FeNOzH3-+ OH- +( NC)&FeNOz4-+ HzO (NC)sFeN02-

where (4) is a rapid acid-base reaction and (3) is rate-determining. For k l ~ AH* , = 12.6 & 0.2 kcal/mole and A S * i 0.7 gibbs/mole. At 298°K klz = 0.55 =k 0.01 M-l sec-l. For the equilibrium kza

(NC)6Fen'0~4$ HzO

k82

(NC)sFeH2O3-

(3)

(4) =

- 17.5

+ NO$-

(2)

AH' = $12.3 =k 0.2 kcal/mole, ASo = 25.2 =t0.7 gibbs/mole, and AGOzss = 4.81 kcal/mole.

KZ3,at 298OK is (3.0

& 0.1) X

and A&* = -3.6 f- 3.0 gibbs/mole. - 28.8 5 3 . 7 gibbs/mole.

The equilibrium constant, M-l sec-1 a t 298OK, AHz3* = 21.6 f- 0.9 kcal/mole For k23, (1.4 0.2) X At 298°K k32 = 0.46 0.07 M-l sec-l, AH32* = 9.3 =k 1.1kcal/mole, and A&* =

+

Introduction Cambi and Szegol in their investigation of the reaction between pentacyanonitrosylferrate(I1) (nitroprusside), (NC)EFeNO2-, and hydroxide ions reported the existence of the equilibrium (NC)5FeNO2- $ 2 0 H -

e (NC)5FeN0z4- + HzO

(1)

and obtained from spectrophotometric measurements Klz = 0.74 X l o 4 (288'K). Kolthoff and Toren2 reinvestigated reaction 1 and obtained KIZ = 1.0 X l o 6 (298'K). The latter investigators also reported the decomposition of (NC)6FeN0z4- to unspecified products and attempted to correct for this decomposition by extrapolation of their spectrophotometric data t o zero time. Zuman and Kabht reported an equilibrium constant of 0.74 X lo5 a t 293°K.3 We have found in this investigation that the reaction between nitro(1) L. Cambi and L. SzegB, Gam. Chim. I t a l . , 68, 7 1 (1928). (2) I. M. Kolthoff and P. E. Toren, J . A m . Chem. Soc., 711, 1197 (1953). (3) P. Zuman and M. KabLt, Chem. L i s t y , 4 8 , 358 (1954).

+

prusside and hydroxide ions is not as simple as has been supposed and that in fact there are two important equilibria established in this system, the first being reaction 1 and the second being

+

(NC)GFeNO~4- HzO

k23 e (NC)5FeOH~3+ NOZksz

(2)

We report here the results of a complete thermodynamic and kinetic investigation of reactions 1 and 2 . Experimental Section Reagents.-Solutions were prepared from J. T. Baker reagent grade chemicals which were used without further purification; KC1 (99.9%), NaNOz (98.9%), NaOH (98.7%), NaZFe(CN)bNO.2HzO (99.6y0), and NaCl (99.5%). Stock nitroprusside solutions were stored in the dark and all operations were carried out in darkened rooms to prevent any light-catalyzed react i o n ~ . It ~ was found that the presence of oxygen had no effect on the measurements made. Instruments.-A Beckman Expandomatic pH meter and low (4) 0. Baudish, Science, 108, 443 (1948)

j74

JAMES

H. SWINEHART AND PETERA. ROCK

Izorganic Chemistry

The total ionic strength was kept a t 1.00 with KC1. The extinction coefficients used in analyzing the absorbance a t 4000 A were: NOz- (1.0); (NC);FeN02(20.5); (h-C)sFeN024-(3.075 X l o 3 M-lcm-l). The solutions were assumed sufficiently alkaline in all cases so that only unprotonated species are of significance (pKa = 3.29 for HN02(aq) and pK, = 4.17 for HFe(CN)63-). 5 , 6 Each value of K12 a t a given temperature mas an w average of five determinations over the concentration m & 1.0 ranges previously indicated. At 298°K the values obtained for K12were: 1.2, 2.1, 1.4, 1.5, and 1.4 x lo6, which yields an average value of (1.5 i. 0.3) X lo6. The values of K D obtained as a function of temperature were: (4.2 1.0) X l o 6 (288°K); (1.5 0.3) X 0 320 350 400 450 550 lo6 (298'K); (0.69 f 0.12) X lo6 (308°K). On reWAVELENGTH ( m p ) cycling the temperature absorbances were found to check the previously obtained values within 5%. From Figure 1.-Spectrum of ( S C ) E F C S O ~a*t-p = 1.00 (SaC1). a plot of log K12 8s. T-l we compute for reaction I : AH" = -18.2 f 0.9 kcal/mole, = -8.43 kcal/ sodium ion error E-2 glass electrode were used for pH measurements. Spectrophotometric data were obtained with an Applied mole, and AS" = -26.1 ==I 3.0 gibbs/mole. These Physics Corp. Cary Model 14, recording spcctrophotometer, thermodynamic data are based on the arbitrarily equipped with a variable temperature cell block. The accuray chosen hypothetical 1 Jl [in = 1.00 (KCl)] standard of the temperature measurement was rir0.5". state. Results and Discussion The equilibrium constant K23for reaction 2 was also determined as a function of temperature. Solutions Thermodynamic Data.-Aqueous solutions of having a total ionic strength of 1.00 (NaCl) which were sodium nitroprusside to l o F 4 M) prepared with 0.2 ,I4 in XaOH, in which the total nitroprusside conan ionic strength of 1.00 (using either NaCl or KC1) were found t o obey Beer's law. If these same solutions [(NC)ZF~OH~~-]) centration @e., [(YC)jFeNOZ4-] to 8 X l o p 4 M, and in which n-as varied from 1 X are made alkaline with NaOH (0.1-0.5X to ensure &l, the NaNOz concentration ranged from 0 to 1 X complete conversion of nitroprusside t o the nitritowere used in the determination of Kza. The extinction pentacyanoferrate(I1) complex) the total absorbance coefficient of the aquopentacyanoferrate(I1) ion was is no longer directly proportional to the total nitrocm-l a t 4000 A.' prusside concentration. However, the absorbance for taken as 1.09 X l o 3 The value of KZ3obtained a t each temperature was a given nitroprusside concentration is reproducible and an average of five determinations within the concentrashows no apparent tendency to change in the course of tion ranges previously indicated. At 298°K the values several hours. It was first observed in a preliminary which kinetic run that the "decomposition" of ( N C ) S F ~ N O ~ ~for - K23were: 3.0, 3.0, 3.2. 2.8, and 3.0 X The yields an average value of (3.0 i. 0.1) X could be prevented by carrying out reaction 1 in the values of KZ3obtained were: (1.1 i. 0.1) X lop4 presence of excess nitrite ion, and i t mas subsequently (283"K), (3.0 i. 0.1) X 10F4(298"K), and (3.8 + 0.2) found that alkaline nitroprusside solutions containing x l o p 4 (308°K). The reproducibility of the observed added excess nitrite ion obey Beer's law. These obabsorbances on cycling the temperature back to 25 " servations imply that reactions 1 and 2 provide a satiswas better than lyO. From a plot of log K23 DS. P1 factory explanation of the behavior of alkaline nitroTve compute for reaction 2 : AH" = 12.3 i 0.2 kcal/ prusside solutions. The visible absorption spectrum of the species (NC)jmole; AGo298 = +4.81 kcal/mole, and AS" = +%.a + 0.7 gibbs/niole (in p = 1.00 (IClaCl), standard state). FeN024- obtained in the presence of excess nitrite I n order to check the assumption that reaction 2 is to 8.00 X 2 1 4 nitroprusside, 0.20 dl (1.00 X NaOH, 4.00 x iV1 NaN02,p = 1.00 (NaCl)) and correct as written and that the reaction does not incorrected for the absorbance of nitrite in this region volve hydroxide ion the absorbance of a solution 4.00 X 10-4 +If in nitroprusside ~ o a staken a t four different by means of a blank is given in Figure 1. The equilibrium constant for reaction 1 was deterhydroxide concentrations and the results obtained were : 0.1 AT,0.781;0.2 M , 0.781; 0.3 AT,0.782; and 0.5 M, mined from the absorbance a t 4000 A in solution, for 0.773. The last value may indicate the onset of the which the "total" nitroprusside concentration (i.e., formation of the (NC);FeOH4- species. At any rate, [ (NC)jFeN02-] [(iVC)~Fe;P\;102~-]) was varied from X and the hydroxide ion concen2 X lo-* to 4 X we conclude on the basis of these data that hydroxide to 3 X M . All tration was varied from 2 X solutions were made 3.00 x lo-? Jf in IzjaNOz, which is ( 3 ) L. G. Sillen and A. E. Martell, "Stability Constants of hletal-Ion Complexes," T h e Chemical Society, London, 1961, p 3 2 . sufficient to ensure the presence of no more than (6) J. Jordan and G. J. Ewing, I J Z O T chi^., ~. 1, 587 (1962). (7) S. Iimori, Z . d i t o i g . Allgem. ClZeiJz., 167, 145 (1927). about 1% of the species (NC)3FeOH23- (vide i n f ~ a ) . 3.0

*

*

+

+

Vol. 5 , No. 4 , A;bril 1966

PROPERTIES OF THE AQUEOUS HYDROXIDE-NITROPRUSSIDE SYSTEM 575 TABLEI EQUILIBRIUM AND THERMODYNAMIC DATA 1~ = 1.00 (KaCl)

Equilibrium

K288

AH', kcal/mole

A S o , gibbs/mole

(KC)jFeNOt.20He (NC),FeX0z4- HjO (NC)5FeN0z4- HzO S (NC)jFeS0z4KOZ-

( 1 . 5 1 0 3) X 1Q6

-16.2 i 0.9

-26.1 f 3.0

-8.43

(3.0 10 . 1 ) x 10-4

+ 1 2 . 3 10 . 2

+25.2 f 0 . 7

+4.81

+

+ + +

ion is not involved in the reaction to any appreciable extent under our conditions. Discussion of Thermodynamic Data.-Table I summarizes the thermodynamic data obtained. The large difference between Klz, (1.5 f 0.3) X 106 a t 298"K, obtained in this investigation and values determined by other workers can be ascribed either to the failure of these workers to recognize the importance of reaction 2 I v 3 or to obtain an accurate value for the extinction coefficient of the species (NC)5FeN0z4-by means of their extrapolation procedure. Kinetic Data.-The kinetics of both equilibria 1 and 2 were followed a t 4000 A, where (NC)5FeN0z4-is essentially the only contributor to the absorbance. When (NC)bFeHzO3- was present the necessary corrections were made. The rate law for the reaction of hydroxide with nitroprusside was found to be first order to both of the reactants. When nitroprusside was treated with excess hydroxide the pseudo-first-order plots were linear over a t least two half-lives and the half-life was inversely proportional to the concentration of hydroxide and 5 X 10-1M . At 286OK k12 as between 2.5 X calculated from In 2/tl/,(OH-) equals 0.21, 0.20, and 0.20 M-I sec-l when the excess hydroxide concentraand 2.5 X M, tion equals 5.0 X 1O-I) 1.0 X respectively, and the initial concentration of nitroM. All runs were carried prusside equals 2.5 X out in excess NO%- to prevent the aquation of (NC)5FeN0z4-. If the mechanism is assumed to be

+ OH- e (NC);FeN02H3(NC)sFeNOzH3- + OH- e ( K C ) & F ~ K O Z+* -HzO hi2

(iYC)5FeN02-

(3)

(4)

where step 4 is rapid compared to step 3, the values of klz in units of M-l sec-l determined as a function of temperature are: 0.20 It 0.01 (286"K), 0.55 f 0.01 (298"K), and 0.92 =t 0.01 (306'K). The values of AHlz* and AS12* computed from the In klz/T vs. 1/T plot were 12.6 f 0.2 kcal/mole and - 17.5 10.7 gibbs/ mole. The kinetics of equilibrium 2 was studied a t high base strengths so that the (NC)jFeN024- complex formed rapidly and the aquation reaction (2) could be studied conveniently. The disappearance of (NC)jFeNOZ4-was followed a t 4000 A. It was assumed that (NC)sFeN024-and (NC)5FeH203- were the only contributors to the absorbance. Pseudo-first-order plots of the disappearance of (NC)jFeN024-were obtained, which were independent of hydroxide (0.25 M and 0.5 M ) and showed linear behavior over a t least one

A C o ~ @ kcal/mole s,

TABLE I1 TYPICAL KINETICDATAFOR AQUATIONOF (NC)sFeN0z4p = 1.00 (NaCl), T = 298°K 1041(NC)nFeN02-]initial.

M

a

(OH-I, M

104k~3,~

A4 -1 sec -1

2.50 0.50 1.25 0.50 2.50 0.25 1.25 0.25 Average of three runs and calculated from k 2 3

1.5 1.3 1.4 1.5 = In 2/ti/,[Hz0]

half-life. Table I1 summarizes typical kinetic data taken under various concentration conditions. At the lowest temperature (298OK) the plot became nonlinear after one half-life as might be expected from the observed decrease in K23 with decreasing temperature (see Thermodynamic Data). Values of kza in units of M-l sec-l determined a t various temperatures (316"K), (6.6 =t0.5) X are: (10.9 f 0.5) X (311°K), (3.2 f 0.2) X (306"K), and (1.4 =t0.2) X l o w 4(298°K). AH23* is 21.6 f 0.9 kcal/mole and AS23" is -3.6 3.0 gibbs/mole. The activation and A&z* for k32 using the thermoparameters aHz3* dynamic A H " and A S o are 9.3 1.1 kcal/mole and -28.8 f 3.7 gibbs/mole. Discussion of Kinetic Data.-Table I11 summarizes the kinetic data and activation parameters obtained. The entropies of activation for kl2 and k 3 2 are in agreement with those predicted purely on the basis of electrostatics. Using AS*,1 = - 10212~,where AS*,1 is the electrostatic contribution to AS* and 21 and Z 2 are the charges on the reactants,* AS*,1 are -20 and -30 gibbs/mole for klz and k32> respectively. It is interesting to speculate as to the structure of the species involved in the mechanism represented by eq 3 and 4. On the basis of Mossbauer effect studies of nitroprusside it can be concluded that the NO group is bound to the iron as NO+.9 The most reasonable mechanism involves the attack of a hydroxide a t the nitrogen forming (NC)5FeI1N02H3- in the rate-determining step. The second hydroxide removes the proton from the complex in a rapid acid-base reaction yielding (NC)bFeN0z4- and water. The reaction of nitroprusside with bisulfide apparently involves a similar mechanism. It appears from stop-flow experiments that the rate of formation of the colored species initially formed in the reaction is first order in both of the reactants. The entropies of activation for the SH- and OH- reactions are nearly

*

(8) A A Frost and R. G. Peaison, "Kinetics and Mechanism," 2nd ed, John Wiley and Sons, Inc , New York, N. Y . ,1962, p 145. (9) J Danon, J Chem. Phys , 41, 3378 (1964).

576 C. F. WEICKAXD FREDBASOLO

Inorganic Chemistry

TABLE I11 KINETICDATAA N D ACTIVATION PARAMETERS, Reaction

+

A H * , kcal/mole

AS*, gibbs/rnde

0.55 i 0.01

12.6 i: 0 . 2

-17.5 i 0 . 7

+ +

+ 0.9

- 3 . 6 i 3.0

+

kx

(NC)sFeN024- H20 -+ (NC)jFeOHa3NO2-

a

1.00 (NaC1)

ki?

(NC)jFeK02- OH( NC)jFeX02H3-

+

p =

Rate constant, Ai'-1 sec-1 (208'K)

(1.4 i0.2)

x

kw

21.6

KO2(SC)ZF~OH~~0.46 i O.OTa 9 . 3 =k 1.10 ( N C ) ; F C N O ~ ~ -Ha0 Values obtained from activation.parametei-s for k23, and A S " and A H " for equilibrium 2.

+

--f

the same and the enthalpy of activation is about 5 kcal/mole smaller for the bisulfide reaction. A complete kinetic and thermodynamic investigation of the system is presently being carried out. From the data obtained it is possible to comment on other reactions in which nitroprusside takes part. CambilO has made observations on the reaction betmeen nitroprusside and acetone. It is clear from our preliminary experiments that acetone does not react with the products of the hydroxide-nitroprusside reaction, (Pu'C)sFeN024- and (NC)sFeH203-. Therefore (10) L. Cambi, Cizem. Zeizt., I, 1756 (1913); t b z d , 11, 1109 (1914).

-28.3 i 3.5iL

the enolate form of acetone must be competing with hydroxide for the site on nitroprusside, the product in the acetone-nitroprusside reaction being (NC)5FeNO(C3H60)3-. This species can then aquate as (NC)sFeN024- does in the hydroxide-nitroprusside reaction. In general base-nitroprusside reactions will involve equilibria of the type in reaction 1, and the species formed will depend on the competitive equilibria between the bases in the system for the site on nitroprusside. Reactions between bases and nitroprusside may provide an interesting route in the preparation of some unique compounds.

COSTRIBUTIOS FROM THE DEPARTXENT O F CHEMISTRY, NORTHWESTERN 'GSIVERSITY, EVANSTOX, ILLISOIS

The Aqueous Solution Chemistry and Kinetic Behavior of a Pseudo-Octahedral Complex of Gold(II1) BY C. F. WEICK1 ASD F R E D BASOLO

Receioed October 27, 1965 The pseudo-octahedral complexes, [hu(Etadien-H)X]PFs2 ( X = C1, Br) have been synthesized and their aqueous solution chemistry investigated. Evidence is presented for hydrolytic and acid-base equilibria. In addition, rate constants for the reaction of [Xu(Etadien-H)Cl] with Br- and OH- and half-lives for the isotopic exchange of this substrate and its conjugate acid with C1- are reported. The rate of chloride ion replacement is almost independent of the nature and the concentration of the reagent. Other reagents such as Na-, I-, and SCX- replace not only the chloride ion but also the triamine t o yield complexes of the type AuXd-.

Introduction

cliff erence in behavior between octahedral and squareplanar complexes is due in part to steric considerations. Because it behaves like an octahedral complex, the complex was designated a pseudo-octahedral complex. Investigations of some reactions of [Au(dien)C1I2 in aqueous solution show that these proceed almost entirely by the path that is first order in reagentG Because of this strong dependence of the rate of reaction on the entering ligand, it seems of interest to test the importance of steric factors using the analogous sterically hindered system. This paper reports the results of (1) Union College, Schenectady, N. Y . such a study on the complex ions [Au(Etddien-H)Cl]+ (2) dien = NH?CHUCH?NHCH~CH?I\"?, Et;dien = (C2Hs)2NCH~CHniYHCHCHzN(CzHa)z, Etadien-H = ( C ~ H ; ) Y ~ C H ~ C H ~ N C H ? C H ~ N ~ and ?. [Au (Et4dien)C1]+.

The rates of substitution reactions of square-planar metal complexes are usually dependent on the nature and concentration of the reagent. Octahedral complexes in aqueous solution generally react a t rates that do not depend on the entering ligand, except for anation reactions. The square-planar, lowspin, d8 complex [Pd(Et4dien)Cl]+has been found to react like an octahedral substrate.G This anomalous result was ascribed to the structure of this ion and suggests the

f

( 3 ) F. Basolo and R. G. Pearson, P i o g v . I w o i g . Chewz., 4, 381 (1062). (4) F. Basolo a n d I